I will protect your pensions. Nothing about your pension is going to change when I am governor. - Chris Christie, "An Open Letter to the Teachers of NJ" October, 2009

Sunday, May 22, 2016

Charter School Realities: Morris, NJ

I was fortunate to participate in a great academic conference at Rutgers this week: Education Reform, Communities and Social Justice: Exploring the Intersections. I'll try to get to some of my impressions later, but for now I want to thank Julia Sass Rubin for inviting me and congratulate her and her staff on doing such a wonderful job (looking forward to next year!).

One of the presenters was Paul Tractenberg of the Rutgers School of Law - Newark. Paul was a key player in the landmark Abbott v. Burke cases, which led to New Jersey's overhaul of its school financing system (sadly, the state has recently retreated from its commitment to funding equity). 

Paul Tractenberg, Rutgers Conference on Education Reform, 5/20/16
photo courtesy of Sarah Tepper Blaine

Paul's presentation was about the Morris, NJ school district, a textbook case of what happens when the courts order school desegregation. Paul wrote about Morris back in 2013:
If we could get beyond our fetishistic attachment to home rule, there are many ways to consolidate districts, either on an individual or statewide basis. Examples of both abound. The Morris School District was created in 1973 out of the adjacent Morristown and Morris Township districts, one increasingly black and lower-income, the other overwhelmingly white and middle to upper income. It was created primarily for racial balance and allied educational reasons. 
Despite initial start-up issues, 40 years later the Morris School District is an amazing success story. It may be the most racially and socioeconomically balanced district in the state, it sends 93 percent of its students on to higher education, and it is widely considered to have been primarily responsible for Morristown’s ability to flower as the state’s leading county seat.
Yet few New Jerseyans are even aware of the existence of the Morris School District, let alone its unique history. By the way, other urban districts sought to follow in Morris’ footsteps in the early to mid-1970s, and again in the mid-1980s, but they were denied that opportunity. 
The result is that today we have the Plainfield, New Brunswick, and Englewood districts standing in stark contrast to Morris as overwhelmingly minority and low-income districts with huge educational problems and in proximity to surrounding predominately white and upper-income districts that once sent their students to them when the urban districts were themselves more diverse.
One thing Paul pointed out during the panel is that Morristown has become one of the most desirable communities in the state, and its diversity is the primary reason. Walk down South Street on a Saturday night and you'll see that the place is hopping: new restaurants, a vibrant arts scene, lots of retail. People call it "Mo-Town" these days.*

During Q-&-A, someone Darcie Cimarusti, aka Mother Crusader, asked about the charter school in the area, and its impact on desegregation efforts. Paul expressed some concerns, but didn't go into details, saying he hadn't looked at the issue carefully yet.

Maybe I can help.

Last year, Unity Charter School applied for an expansion -- its third such application in five years. As is the case for all charters in NJ, expansions are granted solely at the discretion of the Commissioner of Education, David Hespe. The commissioner was clearly influenced by community opposition to Unity's expansion, led by a grassroots local organization, Morris Cares About Schools, and rejected the expansion.

I was following this saga and had planned to do an analysis similar to the ones I did for Red Bank and East Brunswick/Highland Park's charter school expansion applications. But when Unity's application was rejected, I figured I'd just move on.

Paul's presentation made me rethink my decision. Given Unity's tenacity, it's almost certain they will try yet again to get an approval for expansion. So let's get the data, courtesy of the NJDOE, on the record.

- Unity Charter School enrolls fewer free lunch-eligible students proportionally than the Morris School District.

Over the past five years, Unity has consistently served fewer students in economic disadvantage than its hosting public school district.

- Unity CS enrolls more white students and fewer Hispanic students proportionally than the Morris School District.

Year after year, Unity enrolls more white students and fewer Hispanic students than the Morris public schools.

-Unity CS enrolls a similar percentage of special needs students as the Morris School District; however, Unity's special need students have learning disabilities that are significantly less costly than Morris's special needs students.

In 2014, Unity CS has a district special education classification rate of 16.9 percent; Morris's rate was 18.8 percent. That difference is better than most comparisons between charters and their host districts in New Jersey. However:

According to a report commissioned by the NJDOE, Specific Learning Disabilities (SLD) and speech disabilities (SPL) are "low-cost" disabilities. Others -- including autism, emotional disturbance, visual impairment/blindness, mental retardation, etc. - are "moderate" to "high-cost" disabilities.

More than 70 percent of Unity's students have low-cost disabilities; the percentage is significantly less for Morris SD. Keep in mind that when funds are distributed to charters, they get more funding when taking more special needs students; however, the type of disability is only disaggregated by speech/non-speech. In other words, a charter gets as much for a student with an SLD as they get for a student with a traumatic brain injury. But charters don't generally enroll the students with higher-cost needs. Morris SD is likely paying a stiff fiscal penalty for enrolling the special needs students Unity does not enroll.

- When accounting for student population differences, Unity CS does no better than Morris SD on test-based outcomes. In fact, in many cases, Unity does worse than its host district.

As this article notes, Morris flat out beats Unity on several measures of proficiency. But let me take this further. What I've done here is use a simple regression model to adjust average school-wide scale scores on the 2015 PARCC exams:

ScaleScore = f(pctFreeReducedPriceLunch, pctSpecEd, pctLEP)**

This is just saying that I've taken every school's score in New Jersey and compared them all to find out how a school's percentage of free or reduced price lunch, special education, and limited English proficient students affects its average test scores in math and English language arts (ELA) at each grade. I then plotted these adjusted scores against the school free lunch percentage.

Here, for example, is how Unity stacks up against Morris's elementary schools in Grade 3 ELA when adjustments are made for student characteristics.

There's really no difference. But look at Grade 5 ELA:

In this model, a 10 point increase in a school's free or reduced-price lunch rate leads to a 4 point drop in its average scale score on the PARCC. Adjusting for that and other student characteristics: Unity is faring considerably worse than any of Morris's schools that report this test score.

Here's the same grade in math:

The differences in adjusted scores differ by grade and test, so I've put them all below. What's clear is that Unity Charter School can't make any case that it is getting results any better than Morris's public schools; to the contrary, their performance on many measures is considerably worse.

The state also measures "growth" in the form of Student Growth Percentiles (SGPs). I have my issues with SGPs, but it's worth looking at the adjusted scores (I use the 2014 SGPs here, as I haven't yet brought the 2015 SGPs into my databases).

Morris SD is subsidizing Unity to the tune of $1.3 million, an increase of $300K just this past year. That seems very hard to justify given these results.

Test scores, of course, are not the only measure of a school's effectiveness. I would, in fact, argue there are many other things we should look at when judging a school's worth. There may be things Unity Charter School is doing that benefit their students but have nothing to do with tests. Good for them -- but let's also be clear that the "choice" Unity's families enjoy comes at a heavy price.

Morris has worked hard as a community to integrate its school district; charters that enroll different student populations threaten that work. The fiscal consequences of having redundant systems of school management can keep necessary resources from reaching students. So when a charter school isn't really providing a "better" education for a community's children after paying this price... what's the point?

Unity's charter is up for its five-year renewal in 2017. Commissioner Hespe should ask himself some hard questions when considering whether Unity is really bringing value to the good people of Morristown and Morris Township. I don't doubt Unity is full of dedicated, hard-working educators who care about their students. I applaud Unity's students for their effort and success.

But is the cost of "choice" really worth it?

See you on the green.

Here are the other adjusted scores:

*A couple of years ago, I saw the great jazz guitarist Pat Metheny play at Mayo Center, and he said it was one of his favorite places to perform: "When you come here, you get all the New York City hipness without the New York City attitude."

** Robust standard errors (N is between 1312 (Grade 3) and 713 (Grade 7)) with typical colinearity checks (VIFs). Free lunch and LEP are from that same year (2015); special education is a three-year average from 2012 to 2014. FL and SpecEd are significant at the p < 0.01 level. LEP significance varies; I decided to keep it in all models for consistency's sake.

Caveat regressor.

No comments: